Prevention of parastomal hernia: a comparison of results 3 years on

clinical
<strong>Prevention</strong> <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>:
a <strong>comparison</strong> <strong>of</strong> <strong>results</strong> 3 <strong>years</strong> on
Abstract
Parastomal <strong>hernia</strong> continues to be a common and distressing problem for patients with stomas and
research investigating prevention strategies is scant. In March 2005 Thompson and Trainor reported that
the introduction <strong>of</strong> a prevention programme for 1 year following stoma formation surgery significantly
reduced the incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>. This follow-up study strengthens the reliability and validity
<strong>of</strong> the findings <strong>of</strong> the first study by confirming a statistically significant reduction in the incidence <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong>s through the introduction <strong>of</strong> a simple non-invasive prevention programme.
Mary Jo Thompson is
Stoma/Coloproctology Nurse
Specialist, Craigavon Area
Hospital Group Trust, Northern
Ireland, and Bernie Trainor is
Stoma/Coloproctology Nurse
Specialist, Daisy Hill Hospital
and Craigavon Area Hospital
Group Trust, Northern Ireland
Key words
■ Parastomal Hernia
■ Stoma care nurse
■ Stoma
■ <strong>Prevention</strong>
strategies
This article has been subject
to double-blind peer review
Parastomal <strong>hernia</strong> remains a common complication
<strong>of</strong> stoma surgery, occurring in 10–50% <strong>of</strong> patients
(Raymond and Abulafi, 2002; Williams, 2003).
Rolstad and Boarini (1996) define a <strong>parastomal</strong>
<strong>hernia</strong> as a bulging <strong>of</strong> <strong>parastomal</strong> skin, indicating
the passage <strong>of</strong> one or more loops <strong>of</strong> bowel through
a fascial defect around the stoma and into the
subcutaneous tissues. A potential area <strong>of</strong> weakness
within the abdominal muscle is created when a
stoma is formed, and this can present problems
for the patient in terms <strong>of</strong> self-image and difficulty
in managing the practical appliances .
Literature review
A literature review undertaken for the initial
study (Thompson and Trainor, 2005) revealed
that there had been little research into the
prevention <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>s. A second
review <strong>of</strong> the literature carried out 3 <strong>years</strong> on
for the follow-up study found there had been
no further research specifically on prevention <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong>tion; however, one review <strong>of</strong> the
information available on incidence, treatment and
prevention <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>s was uncovered
(McGrath et al, 2006).
McGrath et al (2006) found that the incidence
<strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> varied according to the type
<strong>of</strong> stoma formed. Reported rates <strong>of</strong> development
<strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> in the literature varied
widely: Harris et al (2003) reported the lowest
incidence at 7% (n=354); Lala et al (2002) and
Arumugam et al (2003) both reported rates
<strong>of</strong> 16% (n=55 and n=97, respectively); and
Pringle and Swan (2001) reported a rate <strong>of</strong> 20%
(n=112). Raymond and Abulafi (2002) reported
an incidence <strong>of</strong> 10–50%. Limitations <strong>of</strong> these
22 gastrointestinal nursing vol 5 no 3 April 2007

clinical
studies included small samples and inconsistencies
in follow-up and documentation <strong>of</strong> the time <strong>of</strong>
development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>, all <strong>of</strong> which
hinder <strong>comparison</strong> <strong>of</strong> <strong>results</strong>.
Follow-up <strong>of</strong> stoma patients postoperatively is
poorly documented with regard to the time <strong>of</strong>
development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>s. Akman (1962)
reported that 67.8% <strong>of</strong> incisional <strong>hernia</strong>s (n=500)
had developed within 1 year postoperatively.
Bucknall et al (1982) found that 57% <strong>of</strong> abdominal
incisional <strong>hernia</strong>s (n=84) had developed within
3 months surgery.
Factors contributing to the development <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong> reported in the literature are
obesity, sex, age, site <strong>of</strong> the stoma, abdominal
distension and chronic cough (Pearl, 1989;
McGrath et al, 2006). The variables examined
in this study differ. Bucknall and Ellis (1984)
reported the contributing factors to be chest
infection, wound sepsis, male sex and age 60+
<strong>years</strong>. Bucknall et al (1982) found a statistically
significant correlation between wound <strong>hernia</strong>tion
Figure 1. Abdominal exercises following stoma-forming surgery. (Reprinted with
kind permission from Respond Plus)
and the elderly, male sex and obese patients
undergoing bowel surgery.
Aim <strong>of</strong> the study
The aim <strong>of</strong> this follow-up study was to ascertain
the reliability <strong>of</strong> the prevention programme carried
out by Thompson and Trainor (2005). Raised
awareness <strong>of</strong> the potential for <strong>parastomal</strong> <strong>hernia</strong>
development, the introduction and teaching <strong>of</strong>
abdominal exercises and advice to use support
belts while undertaking heavy lifting or heavy
work until 1 year post-surgery were assessed to
ascertain their continued effectiveness in minimizing
the development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>.
Method
The follow-up analysis utilized year 1 data as
the control for <strong>comparison</strong> <strong>of</strong> the effectiveness
<strong>of</strong> the programme as it was in the initial study
(March 2005):
Year 1: A retrospective study <strong>of</strong> patients who
had a new stoma formed between August 2001
and July 2002 to determine the incidence <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong> formation.
Year 2: A prospective study <strong>of</strong> patients who
had a new stoma formed between August 2002
and July 2003 to determine the incidence <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong> formation. These patients
received active education on abdominal exercises
(Figure 1) and were encouraged to use support
belts or girdles to minimize the risk <strong>of</strong> developing
a <strong>parastomal</strong> <strong>hernia</strong>.
Year 3: A prospective study <strong>of</strong> patients who
had a new stoma formed between August 2003
and July 2004 to determine the incidence <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong> formation. These patients
received active education on abdominal exercises
(Figure 1) and were encouraged to use support
belts or girdles to minimize the risk <strong>of</strong> developing
a <strong>parastomal</strong> <strong>hernia</strong>. This arm <strong>of</strong> the study
was utilized to test the reliability <strong>of</strong> the <strong>results</strong>
obtained in year 2. Follow-up was carried out
for one year post-surgery.
As in the initial study, convenience sampling was
used to facilitate the capture <strong>of</strong> all patients who
underwent surgery for stoma formation at the
two centres involved in the study. This method <strong>of</strong>
sampling enabled the maximum number <strong>of</strong> patients
possible to be recruited during the proposed
timescale. Within year 1 (August 2001 to July
24 gastrointestinal nursing vol 5 no 3 April 2007

clinical
2002) 101 patients were recruited, <strong>of</strong> whom 14
died, resulting in an overall sample <strong>of</strong> 87 for this
year. Within year 2 (August 2002 to July 2003)
138 patients were recruited, <strong>of</strong> whom 24 died,
resulting in an overall sample <strong>of</strong> 114 for this year.
Within year 3 (August 2003 to July 2004) 148
patients were recruited, <strong>of</strong> whom 49 died within
the 1-year follow-up and were therefore excluded,
resulting in an overall sample <strong>of</strong> 99.
The prevention programme introduced in
year 2 continued to be used in year 3 and is
outlined below:
1. On discharge, patients were advised to avoid all
heavy lifting for 3 months following surgery.
2. At 3 months postoperatively:
i. Patients were advised <strong>of</strong> the potential risk <strong>of</strong>
the development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>.
ii. Abdominal exercises were taught and
demonstrated to patients (Figure 1). Patients
were advised to undertake these exercises
daily until 1 year postoperatively.
iii. Patients were measured and support belts
or girdles were ordered. Patients were
instructed and encouraged to wear these
while lifting anything heavy or undertaking
heavy work.
3. Patients were monitored for the incidence <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong> for 1 year postoperatively
at regular intervals (3, 6 and 12 months).
The above advice was reinforced at each
clinic appointment.
Results and discussion
A total <strong>of</strong> 387 patients were recruited to the
study over the 3-year period; 87 <strong>of</strong> these died,
leaving 300 patients involved in the study. The
Figure 2. Age distribution <strong>of</strong> study participants over the 3-year period.
age distribution <strong>of</strong> these patients is shown in
Figure 2. It can be seen that the modal age
group for <strong>years</strong> 1, 2 and 3 is 61–80 <strong>years</strong> (n=51,
59 and 53, respectively).
The incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> in year 1
was 28% (n=24) (Figure 3). This is similar to the
rates reported in some <strong>of</strong> the previous studies,
e.g. 20% (Pringle and Swan, 2001) and 10–50%
(Raymond and Abulafi, 2002).
In year 2, following the introduction <strong>of</strong> the
prevention programme, the incidence <strong>of</strong> <strong>parastomal</strong>
<strong>hernia</strong> dropped to 14% (n=16), rising to 17%
(n=17) in year 3 at 12 months postoperatively
(Figure 3).
Analysis <strong>of</strong> the <strong>results</strong> using chi-squared (P≤0.025)
showed a significant difference between year
1 <strong>results</strong> and year 2 <strong>results</strong>, but no significant
difference between year 1 and year 3 <strong>results</strong>, which
suggested unreliability <strong>of</strong> the findings. However,
year 3 patients had been asked about compliance
with the programme at each review appointment,
and closer examination <strong>of</strong> the <strong>results</strong> revealed
that seven <strong>of</strong> the patients reported not having
undertaken the exercise programme or using
support garments for heavy lifting/heavy work
as instructed. When the revised figures for those
following the programme (n=10) were analysed
using chi-squared, a statistically significant difference
between <strong>years</strong> 1 and 3 was found (P≤0.01). This
finding thus reinforces the effectiveness <strong>of</strong> this
programme in preventing <strong>parastomal</strong> <strong>hernia</strong>, and
demonstrates reliability <strong>of</strong> the programme.
Second stage follow-up in May 2005
In May 2005, all the patients involved in <strong>years</strong> 1
and 2 <strong>of</strong> the study were reviewed again to find
out whether any further patients had developed
<strong>parastomal</strong> <strong>hernia</strong>.
Year 1: The incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>
was 49% (n=75 patients; 26 deaths). The mean
time from surgery was 41 months.
Year 2: The incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>
was 30% (n=91; 47 deaths). The mean time
from surgery was 29 months.
When these figures were analysed using chisquared,
the difference between the <strong>years</strong> was
statistically significant. It is important to note that,
because <strong>of</strong> the deaths, a longitudinal follow-up <strong>of</strong>
patients can potentially skew <strong>results</strong>, so these figures
need to be interpreted with that in mind.
26 gastrointestinal nursing vol 5 no 3 April 2007

Stoma type
Analysis <strong>of</strong> stoma type in year 3 patients
showed that there were more patients with
colostomies (n=51) than with ileostomies (n=43)
or urostomies (n=5).
In the initial study, when the distribution <strong>of</strong>
<strong>parastomal</strong> <strong>hernia</strong> was analysed according to
stoma type, the differences were not statistical
significant. This suggested that, irrespective <strong>of</strong> the
type <strong>of</strong> stoma, all stoma patients had an equal risk
<strong>of</strong> developing a <strong>parastomal</strong> <strong>hernia</strong>. However, when
the figures for year 3 in the present study were
analysed, it was found that 13 <strong>of</strong> 17 colostomy
patients had developed a <strong>parastomal</strong> <strong>hernia</strong>
compared with only 4 <strong>of</strong> 17 ileostomy patients;
chi-squared analysis showed these differences
to be statistically significant (P≤0.01), indicating
that there was a correlation between stoma type
and incidence <strong>of</strong> <strong>hernia</strong>tion. Data from the initial
study were therefore revisited and chi-squared
tests were carried out individually for each year.
This time statistically significant differences
were found for each <strong>of</strong> the 3 <strong>years</strong>, reinforcing
the suggestion that patients with colostomies
are more likely to develop a <strong>parastomal</strong> <strong>hernia</strong>
than patients with ileostomies and urostomies
(Figure 4). These <strong>results</strong> support those <strong>of</strong> McGrath
et al (2006) who, while reviewing the literature
on risk factors for <strong>parastomal</strong> <strong>hernia</strong>tion, found
that the incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> varied
with the type <strong>of</strong> stoma formed.
When the distribution <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>s in
<strong>years</strong> 1, 2 and 3 was analysed according to age <strong>of</strong>
the patients it was found that none <strong>of</strong> the patients
aged 40 <strong>years</strong> or less developed a <strong>parastomal</strong> <strong>hernia</strong>
(Figure 5). A possible reason for this may be that
younger patients show better compliance with the
programme or have more active lifestyles.
The literature could be seen to support this
finding, e.g. Bucknall and Ellis (1984) reported
that age 60+ was a contributing factor in the
development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>. It was noted
that the modal age group for patients who
developed a <strong>parastomal</strong> <strong>hernia</strong> was 61–80 <strong>years</strong>
in year 1 and 41–60 <strong>years</strong> in year 2; however,
in year 3 it changed back to the 61–80 <strong>years</strong>.
When these <strong>results</strong> were tested for significance
using the chi-squared test, a value <strong>of</strong> P≤0.05 was
obtained, indicating that the differences were
statistically significant. On closer examination
<strong>of</strong> the data, it was noted that all <strong>of</strong> the patients
in age group 61–80 who had developed a
<strong>parastomal</strong> <strong>hernia</strong> had a colostomy.
Time <strong>of</strong> development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>
The time <strong>of</strong> development <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>
was examined only in the prospective arms <strong>of</strong> the
study (<strong>years</strong> 2 and 3). The retrospective arm <strong>of</strong>
the study (year 1) was not examined because <strong>of</strong>
poor documentation <strong>of</strong> the time <strong>of</strong> development
<strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>. In year 2, 56% <strong>of</strong> patients
developed a <strong>parastomal</strong> <strong>hernia</strong> within 6 months <strong>of</strong>
surgery; in year 3 the timing was almost identical
to that in year 2, which again suggests reliability
<strong>of</strong> the <strong>results</strong>. In <strong>years</strong> 2 and 3, 58% <strong>of</strong> patients
developed their <strong>hernia</strong> within 6 months <strong>of</strong> surgery,
which again is supported by the literature.
These <strong>results</strong> are comparable to those <strong>of</strong>
Bucknall et al (1982), who showed that 57%
<strong>of</strong> abdominal incisional <strong>hernia</strong>s (n=84) had
developed before 3 months postoperatively,
and reinforce the need to endorse the use <strong>of</strong>
abdominal exercises and support garments for
clinical
Figure 3. Incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> in <strong>years</strong> 1, 2 and 3 (Yes = <strong>parastomal</strong>
<strong>hernia</strong> present; No = no <strong>hernia</strong>tion).
Figure 4. Distribution <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> according to stoma type in <strong>years</strong> 1, 2 and 3.
vol 5 no 3 April 2007 gastrointestinal nursing 27

clinical
Figure 5. Distribution <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> according to age in <strong>years</strong> 1, 2 and 3.
heavy lifting from as early as discharge from
hospital instead <strong>of</strong> 3 months postoperatively.
Limitations
The study has limited generalizability because
it was a relatively small sample. In <strong>years</strong> 1 and
2 we suspected that patients’ non-compliance
with the programme may have affected the
<strong>results</strong>. In year 3 we proved this to be correct, by
inquiring about and recording non-compliance
with the prevention programme. When the
non-compliant patients were excluded from
chi-squared analysis, the programme was proven
to be statistically significant in reducing the
incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>.
We used convenience sampling to capture
all patients within the timeframe <strong>of</strong> the study,
in an attempt to ensure that our sample was
representative <strong>of</strong> patients treated within our
catchment area.
Recommendations
A longitudinal follow-up <strong>of</strong> the patients involved in
the study <strong>of</strong> at least 5 <strong>years</strong> should be undertaken
to ascertain whether length <strong>of</strong> time with a stoma
affects the incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>. However,
it should be borne in mind that the <strong>results</strong> can be
skewed by the death <strong>of</strong> some <strong>of</strong> the patients, as
highlighted in the second stage follow-up, making
it difficult to obtain a true picture. Perhaps followup
for incidence <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong> needs to
be carried out according to diagnosis in order to
truly understand the implications <strong>of</strong> <strong>parastomal</strong>
<strong>hernia</strong> development to patients. An example is
if patients are <strong>of</strong>fered a stoma for quality-<strong>of</strong>-life
issues, the risk <strong>of</strong> development <strong>of</strong> <strong>parastomal</strong>
<strong>hernia</strong> may need to be provided for the patient
to truly make an informed choice whether to
proceed with surgery.
The study needs to be repeated with larger
patient numbers in other centres and in other
countries. Nevertheless, the year 3 <strong>results</strong> have
further reinforced the validity and reliability <strong>of</strong>
this study.
Conclusion
This study continues to have a significant impact
on the way that stoma care nurses provide care
and advice on prevention <strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>s.
It remains the first UK study to have shown a
statistically significant reduction in the incidence
<strong>of</strong> <strong>parastomal</strong> <strong>hernia</strong>s through the introduction
<strong>of</strong> a non-invasive prevention programme. The
programme is within the reach <strong>of</strong> all ages <strong>of</strong>
patients and should be recommended for all
patients for at least 1 year following surgery for
the formation <strong>of</strong> a stoma. ■
Akman PC (1962) A study <strong>of</strong> five hundred incisional <strong>hernia</strong>s.
J Int Coll Surg 37:125–42
Arumugam PG, Bevan L, Macdonald L et al (2003) A
prospective audit <strong>of</strong> stomas – analysis <strong>of</strong> risk factors and
complications and their management. Colorectal Dis 5(1):
49–52
Bucknall TE, Ellis H (eds) (1984) Wound Healing for
Surgeons. Baillière Tindall, London
Bucknall TE, Cox PJ, Ellis H (1982) Burst abdomen and
incisional <strong>hernia</strong>: a prospective study <strong>of</strong> 1129 major
laparotomies. Br Med J (Clin Res Ed) 284(6320): 931–3
Harris DA, Egbeare D, Benjamin H, Jones S, Foster ME
(2003) Analysis <strong>of</strong> stomas and their complications in a 10year
period in a district hospital (Poster Abstract). Colorectal
Dis 5(Suppl 1): 69
Lala A, Hamilton J, Daniel T (2002) Audit <strong>of</strong> stoma care
services in a district general hospital (Poster Abstract).
Colorectal Dis 4(Suppl 1): 63
McGrath A, Porrett T, Heyman B (2006) Parastomal <strong>hernia</strong>:
an exploration <strong>of</strong> the risk factors and the implications. Br J
Nurs 15(6): 317–21
Pearl RK (1989) Parastomal <strong>hernia</strong>s. World J Surg 13(5):
569–72
Pringle W, Swan E (2001) Continuing care after discharge
from hospital for stoma patients. Br J Nurs 10(19): 1275–88
Raymond TM, Abulafi AM (2002) Parastomal <strong>hernia</strong> repair
– a novel approach. Colorectal Dis 4(Suppl 1): 64
Rolstad BS, Boarini J (1996) Principles and techniques in the
use <strong>of</strong> convexity. Ostomy Wound Manage 42(1): 24–34
Thompson MJ, Trainor B (2005) Incidence <strong>of</strong> <strong>parastomal</strong>
<strong>hernia</strong> before and after a prevention programme.
Gastrointestinal Nursing 3(2): 23–7
Williams J (2003) Parastomal <strong>hernia</strong>. The ia Journal
28 gastrointestinal nursing vol 5 no 3 April 2007